We found unprecedentedly high abundances of microbially produced CH 4 in the anoxic deep waters of Lake Untersee, an oligotrophic, perennially ice-covered Antarctic freshwater lake. The maximum CH 4 concentration (approaching 21.8 Ϯ 1.4 mmol L Ϫ1 ) is one of the highest observed so far in a natural aquatic ecosystem. Although surficial lake sediments are the predominant source of CH 4 in Lake Untersee, methanogenesis occurs also within the anoxic waters. Radiocarbon labeling experiments show that H 2 /CO 2 reduction is the predominant methanogenic pathway (90-100%) both in the sediments and the water column, whereas acetate is only a minor CH 4 precursor. This result is consistent with the stable carbon isotope fractionation between coexisting CH 4 and CO 2 . In the water column, CH 4 is partly consumed by both aerobic and anaerobic microbial oxidation as evidenced by CH 4 concentration patterns, stable isotope analyses ( 13 C, 2 H), and 14 C-CH 4 assays. Dissimilatory sulfate reduction also occurs and peaks at 84 m water depth (1.83 mol SO 4 L Ϫ1 d Ϫ1 ). Intense methanogenesis in surficial lake sediments, diffusion of CH 4 from sediments to the water column, additional CH 4 production in the water column, gross CH 4 production higher than CH 4 consumption, and lack of mixing because of the permanent ice cover cause the exceptionally high CH 4 concentration in the lake. Our studies demonstrate that H 2 /CO 2 reduction may sometimes be the major pathway of methanogenesis in low-sulfate freshwater environments even at low temperatures. This pathway is obviously more important in Antarctic lakes than hitherto assumed.As an environmentally important greenhouse gas, methane (CH 4 ) plays a significant role in the global climatic system. Studies of CH 4 cycling in various environments is therefore of fundamental interest. Biological processes are the primary source (ca. 80-90%) of atmospheric CH 4 (e.g., Cicerone and
AcknowledgmentsWe thank M. Schwab, D. Schachtschneider, and G. Müller for assistance in the field; K. Weingart, W. Städter, G. Schäfer, and L. Schönicke for technical assistance. H. Kämpf kindly provided SO 4 , NH 4 , and DIC analyses. We thank A. Mackensen for providing the ␦ 13 C-DIC analyses, and P. Harting for providing solubility data. The acetate analyses were done by L. Dulov, Institute of Microbiology, Moscow. The manuscript benefited from critical comments made